Twitter | Pretraživanje | |
Andrew Erskine
Neuro postdoc @ USC. Neurophysiology of sensation and behaviour.
45
Tweetovi
333
Pratim
159
Osobe koje vas prate
Tweetovi
Andrew Erskine 27. pro
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 26. pro
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 14. stu
Odgovor korisniku/ci @KevenLaboy
I think it would depend on what you need to track on the whiskers. I'm sure it would have no problem tracking whisker base, angle etc. Whisker tip contacts with the texture might be more tricky!
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 7. lis
Calling time on my first . Introducing medium-sized field simulator 2019, made from scratch in <72 hours.
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine proslijedio/la je tweet
Andrew Hires 29. kol
Our deep-dive into mouse behavior, to address how mice locate objects with their whiskers was published OA today in . If you are interested in learning more, please see this Tweetorial by the first author, Jon Cheung.
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine proslijedio/la je tweet
БEЛDΛМ 15. lip
Odgovor korisniku/ci @neauoire
Livecoding music video in by .
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine proslijedio/la je tweet
bxv_neurosci 9. ožu
CHIME: CMOS-hosted in-vivo microelectrodes for massively scalable neuronal recordings
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 7. ožu
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 29. stu 2018.
Odgovor korisniku/ci @DiegoGuarin19
Not extensively but it does still work. Analyzing frames can be slow on local CUP but see:
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 28. stu 2018.
Odgovor korisniku/ci @DiegoGuarin19
Yes, 2 rows of 3 whiskers left intact here. Frame rate ~300hz here to get an imaging region that spans all whiskers in 2 planes.
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine proslijedio/la je tweet
Mackenzie Mathis 24. stu 2018.
Odgovor korisniku/ci @TrackingActions
And, lastly... my favorite part: we demo how to use to train a single camera-invariant network on a very challenging problem-> a cheetah in high-speed pursuit:
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine proslijedio/la je tweet
Romeo RACZ 5. stu 2018.
jULIEs: custom neural probes for recording and stimulation at poster LLL48. Get in touch to see how we're supercharging research and get yours today
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 4. lis 2018.
This was with the original DeepLabCut code? I had similar errors running on the cloud / virtualenv for the first time. Adaptations in the dlc-cloudml repo should account for some of them with more explicit import statements
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 4. lis 2018.
Haven't tried yet, training finishes relatively quickly for me but would still be great to see if it can be sped up. Not sure but perhaps just a case of changing the scale tier in the config file?
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 2. lis 2018.
Odgovor korisniku/ci @incisrdrew
I used Google's cloud ML engine for the model training step, which required some adaptation of the original code. Maybe useful for those without access to local GPUs for continuous training:
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 2. lis 2018.
Blown away by how well works. Completely random frame selection + overnight model training resulted in a tracker capable of following multiple whiskers simultaneously across different imaging planes, even during overlap.
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 24. lip 2018.
Odgovor korisniku/ci @incisrdrew
This I think is a kettle for boiling nightmares
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine 24. lip 2018.
Had a great time at today. Check out this pair of rascals.
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine proslijedio/la je tweet
The Crick 31. svi 2018.
New reveals that brain activity in mice during smell-learning tasks changes rapidly, improving odour discrimination and detection due to changes in sniffing. Read the paper from ’s lab:
Reply Retweet Označi sa "sviđa mi se"
Andrew Erskine proslijedio/la je tweet
Mackenzie Mathis 10. tra 2018.
Automated markerless tracking of user-defined body parts using deep learning! Part of our quest to quantify behavior to better understand the brain. Was a true pleasure to work on this with and ! Our preprint:
Reply Retweet Označi sa "sviđa mi se"